any DLP chip

Hello Dr Pavel,

Thank you for using our DLP Forums.
Our DMDs were designed to work with our controllers and FPGAs.
But can you let me know what is your power budget for your application? Perhaps our Pico chipsets may be more what you're looking for.

Regards,
Oscar

Dear Oscar,

thank you very much for your response. The target  average power consumption of the system is in miliWatts. The DLP chip would work in a "pulse" mode with varying FPS, sometimes the image update would be needed very rarely, e.g. once an hour, the maximum is  0.1FPS. We have the Pico modules, but it is not a viable option because of  the chipset. Also, the generation of the video signal costs energy and it is entirely needless.

In my view there is no real reason to use the controller and FPGA provided we will get the control waveforms/protocol. Is it possible ? 

In summary, I see the essence of the  problem in  the lack of the control waveforms/protocol.

Thank you very much.

Kind Regards

Pavel

Dr. Zahradnik

Reliable function and operation of the TI DLP DMD's  require that they are used in conjunction with the controller chip for that family of DMDs

TI does not release the controlling protocols as there is no way to ensure that the DMD will operate reliably per the TI data sheet limits and warranty period without  the defined interaction of the DMD and controller as a system solution

If you can share more information on your intended application we can suggest the best combination of DMD and controller. 

Do you have a target resolution and wavelength you need to operate with?

Dear Carey Ritchey,

thank you for your response. I wrongly assumed that the non-disclosed controlling protocol is a trade secret. Now, I see that it is not a  secret, it is a protection of the DLP chip.

Without to be too self-confident, I think that after my more than 30 years long career in electronics, microelectronics, digital signal processing and reliability, among others in satellite technology, I am able to design and construct a reliable driving circuit for  the DLP chips.  I measured the driving signals in your Pico kit and I really found no stellar problems, if any.

The operating wavelength is between 530nm-595nm. A resolution 640x360 pixels is fine. More than 640x480 is unnecessary. Unfortunatelly, I cannot disclose the anticipated application, but it is not automotive, not life preserving etc.

Thank you very much for consideration.

 

Kind Regards

 

               Pavel Zahradnik

 

 

Prof. Dr. Prof. honor Pavel Zahradnik

Czech Technical University Prague

Dept. FEL CVUT K332

Technicka 2

CZ-16627 Praha 6

Czech Republic

Phone : (420) 2 2435 2089

Fax : (420) 2 3333 9810

zahradni@fel.cvut.cz

The key is the controlling protocol which is still not available.

Dr. Zahradnik

Based on your description it would be best for your look at one of our Advanced Light Control DMD options

DLP4500 with DLPC350 controller is probably the best option based on your lower resolution needs

We have many industrial customers that use DMDs in similar use cases as you describe with long durations between the need to change the pattern on the DMD array

We offer a DLP4500 EVM that you can purchase to experiment with. your can preload up to 48 1-bit images into the DLPC350 controller using the pre-loaded pattern mode

www.ti.com/.../dlplcr4500evm

You can try: loading the DMD array with the desired pattern then place the DLPC350 controller in standby mode to reduce the controller power draw

Dear Carey Ritchey,

thank you for you response. Our collaborating company has experience with the DLPC350. It is not a viable option as it is current hungry when writing to the DLP chip. Moreover, as explained previously, the WXGA resolution of the DLP4500 is unnecessary.
We have an extremely low-power proprietary MCU which we would like to use to meet the power constraints. An optimal solution for us would be the TI's 640x360 or 640x480 DLP controlled by our MCU. Thank you.
Kind Regards

Pavel Zahradnik

Dr. Zahradrik

One other option is the new DLPC3470 controller combined with DLP2010

www.ti.com/.../dlpc3470.pdf

Dear Carey Ritchey,

 

thank you for your reponse. I have consulted the dlpc3470.pdf

The current consumption stated in Sec. 6.5 - Electrical Characteristics over Recommended Operating Conditions

is prohibitevely high even if IDLE enabled for our application. To put in contrast, our light source consumes no more than 35uA. Our low power MCU draws no more than 20uA/MHz.

 

I still mean that the barrier in low-speed low-power applications of TI's DLP chips is represented by the power hungry TI's collateral chips. It seems that they are not designed with low power applications in mind.

 

Sincerely yours

 

Pavel Zahradnik

Dr. Zahradrik

Sorry to hear that you cannot find a solution based on our large number of options

Good luck in your endeavors

Dear Carey Ritchey,

thank you for your efforts despite no viable solution proposed.

The conclusion I will forward to our industrial partner reads as follows:
1. The TI DLP chips cannot be controller by a proprietary ultra low-power HW because of a secret controlling protocol.
2. The TI DLP controllers are power hungry in a quasi/static mode of operation compared to a possible proprietary solution.
3. Because of 1. and 2., the TI DLP chips cannot be used in the anticipated power sensitive application.

Thank you again.

Kind Regards

Pavel Zahradnik


Prof. Dr. Prof. honor Pavel Zahradnik
Czech Technical University Prague
Dept. FEL CVUT K332
Technicka 2
CZ-16627 Praha 6
Czech Republic
Phone : (420) 2 2435 2089
Fax : (420) 2 3333 9810
zahradni@fel.cvut.cz